Conventional Unmanned Aerial Vehicles (UAVs) operate in various environments and terrains. Future UAV teams are envisioned to be highly autonomous, not requiring constant attention from a control base station. These autonomous UAV teams will likely communicate over a radio frequency link with the control base station. If several autonomous UAVs are operating as part of a team, these UAV team members will likely communicate with each other over a communication network as well. These UAV members will likely be required to inform each other of their respective and absolute positions and flight plans so that they don't hit each other and, since these UAV members are operating autonomously, to continually adjust flight plans to react to the environment, the terrain, and to enemy threats. Each UAV member in the team would communicate periodically with other members of the team.
Members of such a team of UAVs would share the limited network data rate capacity with each other, with other teams of UAVs, and with other elements of the battle space. As a mission unfolds, the changing battle and consequent communication needs of each element of the force may change. The individual demands upon the communications network aggregate and may use up all of the available data rates such that the next separate demand would go unfulfilled in the immediate time scale.
When a UAV team enters a battle area, its communication demands may increase as the members collaboratively plan target engagements and flight plans. The members may simultaneously transmit and receive target tracks and flight plans that may rapidly change. The UAV members of such a team further may conduct collaborative sensing and targeting and update common relevant operational pictures (CROPs), C4ISR information, command and control information, etc. Additional demands may also be placed on the team by non-UAV elements of the battle space.
All of these simultaneous communication demands of the team place stress on the communication network of the team by using up the available data rate, or bandwidth. However, a network typically has “quality of service” algorithms to react and reallocate network resources to those UAV members using it the most. When this happens, some of the UAV members will experience increased delays in their message deliveries, lost packets of information, and other types of service degradation.
For example, assume that each UAV member needs a communications channel with a data rate of 2,000 kilobits per second (kbps) to transmit a 500 kilobit image file in 250 milliseconds (500/2,000). Typically, three images will be sent in succession and this would take about 750 milliseconds. Add another 250 millisesconds for various intermediary processing tasks and the entire process of transmitting and receiving the three images may take 1,000 milliseconds, or 1 second. Thus, the communications channel is entirely consumed by the transmission of these three images for 1 second. Enlarge this concept to a team of five UAV members sharing a communications network with a maximum simultaneous capacity of 10,000 kilobits per second. This communications network would be able to support the transmission of five simultaneous three-image sets of files from these UAV members to a base station. In “non-stressful” situations, this example communication network's underlying data rate is sufficient to support all five UAV members.
However, typically other conditions may restrict the available data rate. Environmental conditions such as rain may reduce the data rate. Enemy jamming may reduce the data rate. Other friendly forces may consume the data rate of the same communications network.
Assume these situations occur and the actual data rate available to the UAV team is only 5,000 kilobits per second. The communications network would then handle only two sets of three image files simultaneously. The network protocols would function in a reactive manner to reallocate the data rates to the earliest transmitted files, not necessarily the two most critical transmitted files. Assume that this takes 250 milliseconds.
Consequently, there would be an additional 250 millisecond delay before the communications network reallocates the available data rates. Furthermore, a third, presumed less critical, set of images would still be delayed with that third set possibly being the most critical set.
Such a conventional system is reactive in nature. Services are reallocated subsequent to the overload occurring within the network. At best, there may be a temporary “bubble” of overload before the qualities of service algorithms begin working. Sometimes a UAV using the most available data rate may not be the UAV with the highest priority, or critical mission need. At worst, the degradation of communications may persist for a period that may degrade critical mission effectiveness.